基于EEMD-JADE的单通道EP信号少次提取方法

摘要

The detection and analysis technology of evoked potential (EP) signal is one of the important method in clinical diagnosis of neurological injury and disease.Existing methods can separate EP signal from mix ing signals after stimulation of hundreds times,but lost its instantaneous characteristics.Therefore,few trial extraction of EP signal is the focus of research in the current biomedical signal.This paper proposed a SCBSS algorithm of EEMD-JADE,which can extract EP signal from the observed signals of single channel mixed with EP and EEG signals of only 2 cycles.The simulation result shows the validity of EEMD-JADE compared with the EEMD-PCA-ICA method.The experiment results show that the EEMD-JADE has better performance in extracting EP signal from observed signals of single channel of only 2 cycles.%生物医学工程中诊断神经系统损伤及病变的重要手段之一是检测与分析诱发电位(Evoked Potential,EP)信号.现有的方法可以在混合信号中分离出经过成多次刺激的EP信号,但丢失了其瞬时特性,因此EP信号的少次提取是当前生物医学信号研究中的一个重点问题.给出了一种基于EEMD-JADE单通道盲源分离算法,只利用2个周期的EP与EEG信号随机混合成的单通道观测信号中分离EP信号,仿真实验证明了EEMD-JADE算法的有效性.并与EEMD-PCA-ICA方法比较,结果表明EEMD-JADE算法从2个周期的单通道观测信号中分离EP信号时性能优于后者.